A curative graft-versus-leukemia (GvL) response following allogeneic hematopoietic stem cell transplantation (HSCT) results from the: (1) restoration of normal immunity, thus overcoming leukemia- or treatment-induced immune deficiencies; and (2) recognition of host leukemia-associated antigens and elimination of leukemic cells. A high priority goal in allo-HSCT is to learn how to generate consistent and potent GvL responses with minimal toxicity. In the parent clinical trial to this proposal, we immunize post-allo-HSCT patients who have advanced chronic lymphocytic leukemia (CLL) with a vaccine consisting of irradiated autologous leukemia cells together with irradiated bystander cells expressing GM-CSF. Our prior studies on donor lymphocyte infusion (DLI), an effective immunotherapy for hematologic malignancies, established that when DLI induces lasting remission, this was associated with rapid recovery of T cell neogenesis, and polyvalent B and T cell immunity targeting leukemia-associated antigens. We hypothesize that potent cell-based vaccines may also enhanceGvL by restoring immune competence and inducing CLL-reactive T cells. To test this hypothesis, we will assess the effects of a CLL vaccine on restoration of normal immune competence and induction of T cell immunity targeting tumor-specific antigens more than self- or allo-antigens. In Aim 1, we will examine how the vaccines affect the distribution of reconstituting immune cells, focusing on changes in the activation state, expansion and repertoire of T cells. In Aim 2, we will determine the balance of T cell reactivity against CLL-specific relative to broadly-expressed self-antigens, since whole cell vaccines contain both CLLspecific antigens and broadly expressed self antigens. To develop effective and safe GvL responses, we will ultimately require targeting of antigens with exclusive expression on CLL cells. In Aim 3, to find such antigenic targets of CD8+ T cells in vaccinated CLL patients, we will use two approaches: first, expression libraries to identify T cell targets; second, a novel approach to identify neoantigens using massively parallel sequencing of tumor DNA vs. germline DNA. To determine whether the identified immunologic effects can be attributed to vaccination, and not HSCT alone, we will analyze in parallel a control group of CLL patients undergoing allo-HSCT but without vaccination. Since we are examining the development of immune responses over time following transplant, our proposed studies require dedicated fresh sample collection and processing. By studying immune reconsitution and identifying target leukemia antigens in vaccinated patients, we will have a foundation for developing the next generation of vaccines to stimulate highly specific leukemia-restricted immunity while reducing toxicity against normal tissues.